template<typename T, typename... Args> threadsafe_lockless_init(void* buf, Args...&& args) {
    new (buf) T(std::forward<Args>(args)...);
}
static decltype(threasfe_lockless_init(nullptr, stuff)) init_ptr = &threadsafe_lockless_init<...>; // init before main

void func() {
    static char buffer[sizeof(T)];
    decltype(init_ptr) init = nullptr;
    InterlockedExchangePointer(&init_ptr, init);
    if (init)
        init(buffer, ...);
    // win
}